ResearchTasksPolicy Question AnsweringFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedPolicyBenchDeepseek-R164.34Accuracy11Apr 15, 2026PolicyBench USDeepseek-R166.43Accuracy11Apr 15, 2026PolicyBench ChineseQwQ-32B65.33Accuracy11Apr 15, 2026